Japanese Phonetic Database

CJKI’s Japanese Phonetic Database (JPD) provides IPA phonetic transcriptions that accurately indicate how Japanese words and proper nouns are pronounced in actual speech, as well as pitch accent codes.

To our knowledge, JPD is the first database of its kind which provides both IPA and accent codes covering Japanese vocabulary. It is no doubt the most phonetically accurate Japanese phonetic database available.

Each entry is accompanied by rich phonological information, such as phonemic and phonetic transcriptions with devoicing, pitch accent codes, grammatical information (part-of-speech codes) and semantic classification codes (type of proper noun). In addition, JPD provides multiple allophonic variants to represent the speech of different speakers.

For example, 鏡 ‘mirror’ /かがみ/ in standard Japanese is pronounced [kagami], where [ga] is a voiced velar stop, but has two allophonic variants, [kaŋami], where [ŋa] is a voiced velar nasal, and [kaɣami], where [ɣa] is voiced velar fricative.

After years of development work based on in-depth knowledge of Japanese phonetics and phonology, CJKI delivered JPD to The National Institute for Japanese Language and Linguistics, the largest and most prestigious government organization that conducts scientific research on the Japanese language. Thereafter, JPD was expanded to cover proper nouns such as Japanese surnames. It is playing a critical role in the development of Japanese speech technology by the world’s leading IT companies.